The company behind Wikipedia wants companies to stop scraping data from their website for their AI training needs.